Research on improving CNN model in internet of things data communication computing
To reduce transmission power consumption and solve the problem of insufficient network bandwidth,traditional methods achieve this by reducing data volume and optimizing computational methods.In this paper,an improved CNN neural network model combined with edge computing was proposed,and the simulation experiments verified that it was superior to cloud computing and traditional edge computing in terms of power consumption,queue average and overall performance.Experiments show that the power consumption is about 15 W lower than cloud computing and about 18 W lower than traditional edge computing,and the overall performance was significantly improved.The proposed improved algorithm model enhances the performance of data communication management in the internet of things and solves the high power consumption problem of data acquisition terminals.
edge computingCNN modelinternet of things datacommunication management
NETL
NSTL
万方数据
